ACL RD-TEC 1.0 Summarization of N04-4040
Paper Title:
A LEXICALLY-DRIVEN ALGORITHM FOR DISFLUENCY DETECTION
A LEXICALLY-DRIVEN ALGORITHM FOR DISFLUENCY DETECTION
Authors: Matthew Snover and Bonnie Dorr and Richard Schwartz
Primarily assigned technology terms:
- algorithm
- boundary detection
- classifier
- decision tree
- decision tree classifier
- decisiontree
- disfluency detection
- downstream processing
- error analysis
- identification
- language modeling
- learning
- learning approach
- lexical modeling
- modeling
- parser
- parsing
- pos tagger
- processing
- recognition
- recognition system
- recognizer
- repair
- semantic analysis
- sentence boundary detection
- speaker identification
- speech recognition
- speech recognition system
- speech recognizer
- speech transcription
- speech-to-text
- statistical technique
- statistical techniques
- tagger
- transcription
- transformation-based learning
- tree classifier
- tuning
- word recognition
Other assigned terms:
- annotation
- annotation specification
- approach
- broadcast news
- broadcast news speech
- case
- community
- confidence score
- context words
- conversational speech
- conversational telephone speech
- data consortium
- discourse
- discourse markers
- error rate
- evaluation data
- evaluation set
- feature
- feature set
- human speech
- hypothesis
- language model
- large training
- lexeme
- lexical features
- lexical information
- linguistic
- linguistic data
- linguistic data consortium
- long distance dependencies
- meaning
- measure
- metadata
- method
- parse
- parse structure
- parse tree
- part-of-speech
- pause
- pauses
- pos tag
- process
- prosodic features
- segment boundary
- semantic
- sentence
- sentence boundaries
- sentence boundary
- silence
- speech data
- style
- subtrees
- tags
- technique
- test data
- text
- tokens
- training
- training corpus
- training data
- training set
- transcript
- transcripts
- tree
- trees
- word
- word sequence
- words
